Deep into Data Analytics

Key Moments

Introduction [00:00]: The hosts, Michael Horn and Jeff Selingo, introduce the episode and discuss the annual New York Times Education Issue and the importance of data science skills across various industries.

New York Times Education Issue [29:20]: Michael and Jeff delve into the New York Times Education Issue, discussing its timing, content, and the use of outside writers for the issue.

Value of Higher Education [15:00]: The hosts analyze an article by Paul Tough on the declining faith in the value of college, discussing root causes of this disenchantment, and the role of informal learning through non-accredited institutions, online platforms, apprenticeships, and certification programs.

College Access Index [21:51]: Jeff discusses the New York Times magazine story David Leonhardt and Ashley Wu about the college access index.

Links We Mention

Turning Data into Wisdom

Upcoming Next Office Hour Webinars and Archives

Connect with Michael Horn:

Connect with Jeff Selingo:

Connect with Future U:

Submit a question and if we answer it on air we'll send you Future U. swag!

Transcript

Michael Horn:

Jeff, you have been busy with a bunch of white paper releases, webinars, conferences. And I know, like me, you were also fielding lots of questions and hot takes about the annual New York Times magazine issue that comes out every September on, they call it The Education Issue. It's almost like a rite of passage at this point. I know to expect it every year. Yet, Jeff, somehow I'm always surprised when it comes out. I even think I wrote for it about a decade ago, so you'd think I would know when it comes out. But nevertheless, I don't learn, I guess. But given all that pile up, particularly on your plate, today's episode I want to do is a bit of a catch-up. It's like let's let Jeff clear his desk of all those piles of things that I know you are burning to talk about.

Jeff Selingo:

Yeah, I'm not quite sure it's ever going to be totally empty, just like I'll never get to zero inbox. I'll be lucky to get to four or three figures on that inbox number, as the people who email me probably know. But speaking of email, Michael, I just want to mention we'd love to hear from our listeners, particularly this year. So just to know how to reach us, you can find Michael at Michael B. Horn both on LinkedIn and X and on the web at michaelbhorn.com in addition Substack and following him on Instagram @MHorn1999. You can find me on LinkedIn or Facebook at Jeff Selingo and on Instagram and X @JSelingo, or subscribe to my newsletter, Next, at jeffselingo.com. And you can also reach the podcast using the contact form on our website, futureyoupodcast.com. And while you're on the site, a reminder to sign up for emails from Future U. We'll send you an email update every couple of months on topics we're discussing or what we're hearing from listeners, and you could subscribe to that at futureyoupodcast.com.

Sponsor:

This episode of Future U is sponsored by Ascendium Education Group, a nonprofit organization committed to helping learners from low-income backgrounds reach their education and career goals. For more information, visit ascendiumphilanthropy.org. This episode is brought to you by the Bill and Melinda Gates Foundation, working to eliminate race, ethnicity, and income as predictors of student success through innovation, data and information, policy, and institutional transformation.

Michael Horn:

I'm Michael Horn.

Jeff Selingo:

And I'm Jeff Selingo.

Michael Horn:

Well, given that we've been talking a lot about enrollment and how colleges can create demand, how about this paper that you recently wrote, and we'll link to it in the show notes, about how data analytics has become sort of the new foreign language or maybe the new computer coding, if you will, the code for all of this decade, and a way maybe even to grow enrollment, Jeff?

Jeff Selingo:

Yeah. So Michael, this is part of a series that I'm writing looking at data analytics. As we know, Moneyball, the famous book, but even probably the more famous movie with Brad Pitt, everybody describes something now as the Moneyball of fill-in-the-blank because it's everywhere. And when data-intensive roles first came on the scene in the 1990s, as I learned in researching this paper, they were really largely held by quants, with people with backgrounds in physics and math. And the title of data scientist came along later in 2012. And get this, it was dubbed the sexiest job of the 21st century by Harvard Business Review. But even a decade ago, there really were very few data scientists in terms of titles. There was only 150 postings across the entire US in 2011. But now the need for employees as anybody out there, any of our listeners knows with data science skills, has really grown and well beyond the tech industry.

You have data science needs in retailers and airlines and government agencies, entertainment companies and sports teams. And so in that way, kind of getting value and telling stories from large datasets, it's really now seen as a, as I found in writing this paper, a table stake skills for many jobs. And in many ways, that's what makes it similar to, remember when we were growing up everybody kept saying, "Take a foreign language, take a foreign language," or computer coding was all the rage back a decade or two ago. And what I found was that, as data science has kind of grown more important, of course colleges and universities have responded. They've created hundreds of new programs in data science and analytics both at the undergraduate and graduate level. But as is typical in higher ed, what I found is that they tend to be siloed in business schools.

So what does all this mean for higher ed going forward? Now the push among employers and academics alike is to really stop thinking of data science as that single discipline, or stuck in the silo of business schools. In fact, the Harvard Data Science Review, which is a new magazine that Harvard started in 2019, is that data science is best understood as a collection of disciplines. In many ways, this is democratizing data analytics, which is important if we want those majoring in English or biology or history to really start to think about these as not only a skillset to have, but perhaps even to go into that profession. That's a way to diversify it.

And really what I think this requires is for colleges and universities to really start to think about infusing this, infusing data science instruction across the entire curriculum. I think I've mentioned this before, my eighth grade daughter is very upset with me because I told her she had to take this data science course that her school was offering this semester. Because I just think... She says, "Well, I'm not a math person. I'm not a science person." I just think, though, that it's going to be part of every job in the future. It already is. And so you have to have some tools.

Michael Horn:

Yeah. I agree with you Jeff. And it's interesting. In K-12, of course, there was the coding for all movement starting somewhere around the middle of last decade I think. And then, as you know, I, along with Stephen Dubner of Freakonomics fame and others have pushed us to rethink sort of the Algebra 2 to calculus sequence. Part of the reason, frankly, Jeff, is for folks like your daughter who maybe say, "I'm not good at math." Well, this might be a more accessible way for them to do math that's actually used in the real world as opposed to calculus or Algebra 2, which engineers, it's something like 6% of engineers say they use Algebra 2 in their day-to-day job. So what are we doing? But the reason I think that we continue to do this sequence, and the reason the push for moving away from that sequence to stats or data science or analytics, et cetera, is so controversial is because people say, "Well, higher ed wants to see calculus." And if you-

Jeff Selingo:

Yeah.

Michael Horn:

... move away from calculus, you're going to signal that it wasn't a rigorous curriculum. And so given that feeling in K-12, I'd love to know how you're seeing this play out in the ground in higher ed. Did you find in your research that this movement toward infusing data science instruction across the curriculum, or even in departments, is it actually happening?

Jeff Selingo:

Yeah, it's a great question. Speaking of calculus, by the way, at K through 12, we have to remember that only I think half of high schools even offer it as a course. But yet college admissions officers, by the way, love, love-

Michael Horn:

Yeah.

Jeff Selingo:

... calculus. But in terms of this question about how higher ed is responding on data analytics, they really are beginning to take note. Probably the most prominent example is the University of California at Berkeley. They agreed earlier this year, and the regents there at the University of California system are creating a new college of computing data science and society. It's the first new college at Berkeley in more than 50 years, so I think that says it all. And it's really designed to meet the huge demand across campus. They had 4,200 students who took an intro to data science course last year. That's 13% of the entire undergraduate student population took this course. Also, the other thing about that Berkeley program, and Michael, I think you'll be interested in this because we're always talking about partnerships in higher ed and alliances as maybe a precursor to M&A, is that the Berkeley curriculum is being shared with 89 other higher ed institutions across the country, including six California community colleges and four Cal state colleges.

Michael Horn:

That's fascinating, Jeff, and encouraging because that's a great way to diffuse this, particularly if you don't maybe have the expertise to create a curriculum on your campus because you haven't had a department there. Or maybe there's certain faculty members who themselves maybe are a little bit scared by building this into their courses. And obviously, like you, I really like this idea of democratizing data science for everyone. And speaking of democratizing data science and using it in different ways, I've been fascinated by the data visualizations in the New York Times lately, and we're going to get to talk about one of them in particular in just a minute. So I would think that journalists, maybe, Jeff, maybe they would've been helped by a background in data science?

Jeff Selingo:

Well, I will tell you, if I'm being a journalism, if I want to go into journalism now, you need to know data science because it's really what will distinguish you. And so the question, though, is how. How do colleges do this? They're overwhelmed by the demands on everything that they're doing right now. As we also know, because we talked about this, colleges don't have a lot of resources. So in my paper, I laid out a framework. I laid out a framework of four ways that education and higher education in particular can go. They could go broad, so they could promote an interdisciplinary approach that recognizes there isn't a single approach. For example, they require all undergraduates take a quantitative reasoning or data literacy course, no matter the major. A data literacy course would be pretty easy to stand up. They could go deep. So instead of doing a single analytics degree, they could actually start a whole school within the university, much like Berkeley is doing.

Or they could provide a variety of concentrations and degrees at various levels. So that is kind of a deep focus on data analytics. They could, as I say, go together. They could encourage data science as a complimentary skill. This is a little bit like having everybody take the course, but not only have everybody take a course, but actually give them a micro-credential in data analytics or data visualization and embed them in the degree. And then finally, if the college or university you're not at doesn't do any of this, you could kind of do it on your own. There are so many DIY pathways now to getting a credential in data analytics, and many of them are free that are out there. So there's one of four pathways that you could go on this front.

And just one final thought on this, Michael, before we move on to other things. And this comes from Adam Weinberg, who's the president of Denison University in Ohio. And Denison was probably the first liberal arts college that I could find in the US to start a data analytics major. And they did that back in 2016. And since then, he told me the major has grown to be the sixth largest at Denison. And students there must complete the program in the context of another academic domain, so biology, economics, environmental science, even philosophy. And what he told me was that it's actually easier to recruit students and even faculty than in computer science because it's a much broader discipline.

Michael Horn:

It's an interesting finding. And it's interesting to see liberal arts colleges getting into this field. And as you said, it's a great way to create real meaningful demand for their programs because there's a real premium for that in the labor market. But with that, and having gone through this paper, Jeff, let's take a short break. And when we come back, I want to get your thoughts on what we've been teasing, which is the New York Times magazine issue on education. This episode is being brought to you by the Bill and Melinda Gates Foundation. Today's college students are more than just students. They're workers, parents and caregivers, and neighbors. And colleges and universities need to change to meet their changing needs. Learn more about the foundation's efforts to transform institutions to be more student-centered at usprogram.gatesfoundation.org.

Jeff Selingo:

This episode of Future U is sponsored by Ascendium Education Group, a nonprofit organization committed to helping learners from low-income backgrounds reach their education and career goals. Ascendium believes that system-level change and a student-centric approach are important for our nation's efforts to boost post-secondary education and workforce training opportunities. That's why their philanthropy aims to remove systemic barriers faced by these learners, specifically first-generation students, incarcerated adults, veterans, students of color, adult learners, and rural community members. For more information, visit ascendiumphilanthropy.org.

Michael Horn:

Welcome back to Future U. And as I mentioned at the top, every September the New York Times magazine comes out with its education issue. Maybe we could ask Jeff why they still do this in September when education is a year-round issue. Although then we would probably have to re-look at why we take a summer break. So I don't know, maybe this is just ingrained in all of us. But more concretely, and away from the timing, what did you make of this year's issue?

Jeff Selingo:

Well, in terms of why they still do it, I mean, the New York Times coverage of higher ed is very good. I think they used to do a lot more of it. They do a lot less of it now, I think, in the daily newspaper. I think Ron Lieber, for example, does a lot on financial aid in terms of personal finance. And obviously they still cover the big issues in higher ed, but I get the feeling that their higher ed reporting staff used to be much larger. The magazine itself is really a separate entity within the New York Times.

So I think they're like 60 Minutes. They like to take a big look at an issue, even if the daily news in that case, CBS evening news, is constantly covering it. It's also interesting to me, they use a lot of non-New York Times writers in that issue. So even though they may have experts on staff, they like to use a lot of outside writers, including somebody who wrote the lead piece, who was Paul Tough, who we all know, who's also an author and a magazine writer. And he wrote the piece on value of higher ed. And I'd love to get your take on that piece because, Michael, I have another piece in the magazine that I want to bring up.

Michael Horn:

All right. Well, let's jump to that in a moment. I'll say, it seems like Paul Tough is on an every four years cycle writing for the New York Times education issue as far as I can tell. But his piece was titled Americans Are Losing Faith in the Value of College. Whose fault is that? And Jeff, I thought for those of us in the field, maybe the most striking stat in the piece was that, a decade ago, 99% of Republicans expected their kids to go to college. And I pulled that one out because I think it's easy to forget because it's changed so dramatically today. And he then goes on to capture the trend lines that our listeners will frankly already know well, the declining enrollments, declining enthusiasm and belief in college.

There's not a lot new there, I think, for our audience. But something he did that you don't see everywhere is he compared it to the college going rates in other countries. And he made the point that America is the outlier here in that we have fewer individuals going to college, whereas all these other countries are having more people go to college. And in his words, quote, "Better educated workforce as a result," than we do. And I'll be honest, Jeff, these stats always drive me crazy because it equates time in school with learning, which, as you know, I'm the mastery-based, competency-based learning guy. The time in learning as measure of education attainment drives me nuts, but I also kind of get why we do it. That said, it ignores sort of all the informal learning that's occurring in the post-secondary space, whether it's through non-accredited institutions like Merit America and Marcy Lab School that we've had on the show, or even YouTube or growing apprenticeships or certification programs.

That said, it is what it is. But Paul Tough then basically says, "Okay, what are the root causes of this disenchantment with higher ed?" And the big one he lands on is that a lot of people like to look at college and its ROI through the college wage premium. But he says the better one to look at is the college wealth premium. So not just the boost in earnings, but the boost that also figures in and calculates your expenses and liabilities. Again, I'm not sure that's super new to people in higher ed, but it's certainly a much more accurate picture of the real return on investment. What's interesting is that if you look at the college wealth premium, while the wage premium is held pretty well, the wealth one has dropped starkly for younger Americans. And the hypothesis that he floats from a number of economists is that because, we all guessed it, it's cost. And there's no, I think, real surprise here or anything new.

But it's probably new to a non-education audience, Jeff, which I think is interesting because this is sort of what we're swimming in, the milieu of higher ed. More important to me was when Tough didn't just talk about the college wealth premium not being as big these days, but more that a central reason I think for that is frankly the downside risk of going to college today. That if you factor that on top of the benefits of the college degree, the price of failure is a lot higher than it used to be. And there's still a lot of failure, 40% roughly of students not graduating within six years, say. And Douglas Webber, he quotes him, the former temple faculty member now at Fed Reserve Board as a senior economist. And he has this basically set of calculations that he does that says, "Look, if college is free, you really ought to go because it's a 90 something percent success rate that it's going to pay off and you'll have a great college wealth premium."

But then he goes, basically, "If you're paying $25,000 a year in tuition and expenses, your chances of coming out ahead drops to about two in three." Well, that's a third of folks failing. And then if it's 50,000 a year in college costs, your odds are no better than a coin flip. Now, that's pretty sobering. You might wind up with more than the typical high school grad, but you're just as likely to wind up with less. And then just the last part, Jeff, on this, because he then goes... So there's sort of the economic argument for the declining faith in college. But then he goes to the political one, which, of course, for folks has been so striking just how much a college degree predicts the way you're going to vote at the moment compared to years past.

And Tough talks about the leftward shift on college campuses. And again, it doesn't feel to me like it's anything new, but he lays it on pretty starkly. And I guess, Jeff, the only thing I'd say that I found surprising was that he had a whole bunch of poll data about how Republicans have lost faith in higher ed, but Democrats, they've sort of held steady. And I guess that's not my read on the data. I'm curious if you have a different take on that, Jeff. But I've seen plenty of data showing that the Democrats, a democratic individual, their faith in higher ed has also slipped. This isn't just a Republican thing because of, say, fewer conservatives on college campuses.

To be clear, that may be an issue, ideological diversity and short supply. But it feels like the economic one, from my perspective, and the failure rate is maybe the more important one. And then the last one he just didn't talk about, Jeff, was frankly related to politics, but costs in politics, which is we hear so much chatter about student loans and student loan forgiveness and the rest, I can't help but think that that causes individuals to say, "Well, maybe that's a bad idea. I don't want to get stuck with college loans." So those were my reflections on the piece. I'm curious what struck you maybe there, but also in some of the other articles that were in the magazine.

Jeff Selingo:

Yeah, there's another piece I want to talk about. But just going back to your point on the polling, my take on this also is that the democratic support has also dropped. The bottom hasn't fallen out, like it has among Republicans. But I also think that's how people sometimes identify with political parties these days as well. And really, the Republicans have become the party of no education rather than where we had parties that were a little bit more equal, and even moreso maybe the Republicans 30 or 40 years ago. But we've seen support among Democrats also drop, largely because they think college costs too much. And I think that's something that maybe is bipartisan on both sides.

So it's not just the Republican side that has dropped on education, although I think that has a lot more to do with the culture wars, the value of higher education. I don't think that's as much of an issue on the democratic side. But Michael, the piece I wanted to talk about is the college access index, which David Leonhardt and Ashley Wu wrote about in the magazine. And it's interesting because David Leonhardt has really been kind of a champion of this. He did this college access index about a decade or so ago. I actually think, and I'm not taking credit for this, I think the Chronicle of Higher Education actually maybe did it first, but when the New York Times does something-

Michael Horn:

Yeah, put your flagpole there.

Jeff Selingo:

Yeah.

Michael Horn:

Let's stick to it.

Jeff Selingo:

Yeah, let's stick to it. But when the New York Times does something, it's like nobody else did it before. And I'll never forget that because when he first did that, it was a cover, it was a front page piece. And it put Washington University and St. Louis on the cover and it embarrassed them. At the time, they were the least economically diverse college in the country. Only about 6% of its students receive Pell Grants. And that really embarrassed Wash U. And I know that from talking to the former provost out there who we've had on our show before, Holden Thorp. And they made this a big push to try to diversify economically. And as I said in the most recent one, they were at 16%, still not that high compared to a lot of other colleges, but much higher than 6%. Now, what's interesting is that, over the years, the New York Times has done this regularly. And really what they did was turn the Pell Grant into a proxy for how well you serve low-income students.

And at most of the 286 colleges in the New York Times analysis, the Pell share actually fell over the past decade. And why is that? Well, because it's expensive. And every institution's not Harvard and not Yale and not Princeton. They don't have billions and billions of dollars sitting around where they can use to make up the difference. The Pell Grant is still kind of a tiny fraction of what you need for tuition. And so somebody has to make up the gap on top of the Pell Grant. And you could do it by loans, but loans are not great for people who are of low income.

And so it's just expensive for colleges and universities, and sometimes it's just not a priority. Full-pay students might be more of a priority for them because they need the revenue. And that's also dropping, by the way, as I noted in my newsletter, Next, a few weeks ago, that 89% of all full-pay students who go to liberal arts colleges, this is still a shocking statistic to me. 89% all full-pay students who go to liberal arts colleges, they're enrolled at one of the campuses in the US News and World Report top 50.

So basically every other liberal arts college in the country beyond the top 50, they're only left with 11% of all the full payers out there who go to liberal arts colleges, which basically means that they're giving a lot of discounts out there. Now, there's one other thing I want to bring up in this story because it goes back to something I just said about the proxy for low-income students, because it's not an exact proxy. And I think that's part of the problem with using the Pell Grant percentage. As some higher ed leaders told the New York Times, they just don't think Pell is a comprehensive measure of diversity. For example, it can't distinguish between a student with a household income of, say, 100,000 and one with a million dollars because neither is likely to receive a Pell Grant. Pell Grants mostly go to students from families making under $50,000. That's kind of a rule of thumb.

And there was this line that David Leonhardt had in his newsletter. And the morning it came out, Ron Lieber texted me a screenshot of this because he thought it was a critical point. And I'll just read from it. It said, "Other data from both academic research and the federal government suggests that most colleges with a low Pell Grant share are not in fact enrolling large numbers of students just above the cutoff. Any college that believes it's doing so is free to release data showing the full income distribution of its students." And so because what's happening is that these college leaders are saying, "Well, our Pell Grant numbers might not be good, but we're enrolling a lot of kids in 50 and 60,000 and $70,000 family income. And we don't get any credit for that in this analysis in the New York Times."

Very true. But as Leonhardt said, and as Ron Lieber said in his text to me, "Okay. Well, if you're doing that, then release those numbers." And it's actually one issue that we're going to bring up in a NACAC session, National Association of College Admissions Counselors. Ron Lieber and I are going to be on a panel together and we're recording this before we go to Baltimore to do that. By the time this comes out, that panel will be over. But we're actually going to bring this up as an issue because colleges say, "Well, we're doing actually better on these measures." Well, if you are, then be transparent about it and release the data.

Michael Horn:

Yeah. And just to stay on that because I always like harping on this, as you know. Nothing keeps the colleges and universities from standing behind outcomes themselves. They don't have to just stand behind what's released. And, I mean, obviously I've been on a bandwagon for this, put out audited numbers of the definitions that you think or the data that you think is important. It does strike me how hard it is to do this with the public data that's out there, though. I think about the K-12 food fights about low-income students, which drive a lot of sensational headlines around how many students are in poverty these days. And they're really using free reduced lunch numbers, which are also a very imperfect proxy for actual poverty itself. But gosh, Jeff, NACAC, you've got a busy fall of events. The airplane might see you more than your bed is. It seems like you're traveling constantly.

Jeff Selingo:

Yeah. I have a few high school talks on admissions. I have the big P3, public-private partnership, conference in Denver. I'm also doing two Next office hours, which are named after my newsletter where we do these hour-long webinars. And we're doing these on AI and higher ed. And also just looking beyond AI as something around students, but more about how we can use AI to build courses and also around administrative functions. There's a lot of interest in both of these subjects, and we'll put a link in the show notes to join those office hours or see recordings of it. Michael, it's interesting to me as I'm recruiting people for these office hours, there's a lot of interest in the use of AI and how we could use AI to build courses and help with administration functions. There's just not a lot of expertise.

We did that one episode on AI last year, and I think we're probably going to need to do another. I just got back from Coursera, their headquarters. They did a big future of higher ed conference where AI was kind of all the rage. And there was a lot of questions about how AI can actually help faculty do their jobs. And so that's one of the issues that we're going to look at in this webinar. How can AI be complimentary to faculty jobs rather than competitive? But with that, I think we're going to leave it there for this week. Thank you all for joining us and for your interest in everything in future of higher ed. And we'll see you next time on Future U.

Deep into Data Analytics

Listen Now!

Join our Newsletter

Transcript

Wherever You Listen to Podcasts